-----------------------------------------------------------------------------------------------------------------
      name:  <unnamed>
       log:  /Users/Mark 1/Dropbox/Grant Proposals ICES/ISQ Response/Accepted/replication/replicating table 2.log
  log type:  text
 opened on:  10 Jun 2015, 16:52:03

. use "/Users/Mark 1/Dropbox/Grant Proposals ICES/ISQ Response/Accepted/replication/crescenzi-kadera-2015-ISQ-Tab
> le2.dta"

. do "/Users/Mark 1/Dropbox/Grant Proposals ICES/ISQ Response/Accepted/replication/crescenzi-kadera-table2.do"

. 
. **  Replication file for Table 2 from Crescenzi and Kadera, 2015. 
. **  "Built to Last: Understanding the Link between Democracy and Conflict
. **  in the International System."
. 
. **  This is a response piece. Original Article: Gartzke, Erik and Alex Weisiger. 
. **  2014 ‚"Under Construction:  Development, Democracy, and Difference as 
. **  Determinants of Systemic Liberal Peace,‚Äù International Studies Quarterly 
. **  58(2):130-145.
. 
. ** Original Gartzke & Weisiger Replication Files can be found here:
. ** http://thedata.harvard.edu/dvn/dv/weisiger
. 
. * Analyses in Crescenzi & Kadera was performed using STATA 13.1
. 
. use "crescenzi-kadera-2015-ISQ-Table2.dta"

. 
. set more off

. 
. *MODEL A, also MODEL 5 in G&W*
. nbreg fat1 polave diff1 pcenerg sysdepstate states year, nolog robust

Negative binomial regression                    Number of obs     =        113
                                                Wald chi2(6)      =     147.67
Dispersion           = mean                     Prob > chi2       =     0.0000
Log pseudolikelihood = -203.31828               Pseudo R2         =     0.1938

------------------------------------------------------------------------------
             |               Robust
        fat1 |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
      polave |  -.0379624   .0687888    -0.55   0.581     -.172786    .0968612
       diff1 |   .8686624   .2388337     3.64   0.000     .4005569    1.336768
     pcenerg |  -.9519497   .4173642    -2.28   0.023    -1.769968   -.1339309
 sysdepstate |  -57.01185   21.48791    -2.65   0.008    -99.12739   -14.89631
      states |   .0166502   .0070228     2.37   0.018     .0028857    .0304146
        year |  -.0008306   .0140713    -0.06   0.953    -.0284099    .0267487
       _cons |  -3.496475    25.2956    -0.14   0.890    -53.07494    46.08199
-------------+----------------------------------------------------------------
    /lnalpha |  -3.391006   1.505989                      -6.34269   -.4393225
-------------+----------------------------------------------------------------
       alpha |   .0336748   .0507138                      .0017596    .6444729
------------------------------------------------------------------------------

. 
. 
. *MODEL B, also MODEL 6 in G&W*
. *nbreg fat1 polave polave2 diff1 pcenerg states year, nolog robust
. nbreg fat1 c_polave c_polave_sq diff1 pcenerg states year, nolog robust

Negative binomial regression                    Number of obs     =        186
                                                Wald chi2(6)      =     239.77
Dispersion           = mean                     Prob > chi2       =     0.0000
Log pseudolikelihood = -275.74488               Pseudo R2         =     0.2456

------------------------------------------------------------------------------
             |               Robust
        fat1 |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
    c_polave |  -.0709127   .0663906    -1.07   0.285    -.2010358    .0592104
 c_polave_sq |   .0038366   .0145449     0.26   0.792    -.0246709    .0323441
       diff1 |   .7355649   .2242902     3.28   0.001     .2959642    1.175166
     pcenerg |  -1.127634   .3942849    -2.86   0.004    -1.900418   -.3548498
      states |   .0120197    .005324     2.26   0.024     .0015848    .0224546
        year |   .0106779   .0105728     1.01   0.313    -.0100443    .0314002
       _cons |  -24.41205   18.69449    -1.31   0.192    -61.05258    12.22848
-------------+----------------------------------------------------------------
    /lnalpha |  -3.801908   2.082499                     -7.883532    .2797158
-------------+----------------------------------------------------------------
       alpha |   .0223281   .0464983                      .0003769    1.322754
------------------------------------------------------------------------------

. *reported centered model, results substantively equivalent*
. 
. *MODEL C, Replacing polity average with dempower*
. *nbreg fat1 dempower dempower2 diff1 pcenerg sysdepstate states year, nolog robust
. nbreg fat1 c_dempower c_dempower_sq diff1 pcenerg sysdepstate states year, nolog robust

Negative binomial regression                    Number of obs     =        113
                                                Wald chi2(7)      =     145.94
Dispersion           = mean                     Prob > chi2       =     0.0000
Log pseudolikelihood = -200.17426               Pseudo R2         =     0.2063

-------------------------------------------------------------------------------
              |               Robust
         fat1 |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
   c_dempower |  -30.42639   11.34292    -2.68   0.007    -52.65811    -8.19468
c_dempower_sq |   225.1175   65.63341     3.43   0.001     96.47843    353.7567
        diff1 |   .9078187   .1964485     4.62   0.000     .5227867    1.292851
      pcenerg |  -1.314228   .3704475    -3.55   0.000    -2.040292   -.5881646
  sysdepstate |  -46.23801   20.20751    -2.29   0.022      -85.844   -6.632022
       states |   .0148846   .0075486     1.97   0.049     .0000896    .0296797
         year |    .005679   .0124812     0.45   0.649    -.0187838    .0301417
        _cons |  -14.96358   22.41883    -0.67   0.504    -58.90368    28.97652
--------------+----------------------------------------------------------------
     /lnalpha |  -3.759775   1.920086                     -7.523075    .0035247
--------------+----------------------------------------------------------------
        alpha |    .023289   .0447168                      .0005405    1.003531
-------------------------------------------------------------------------------

. 
. 
. *MODEL D, Replacing polity average with Democratic Community
. * Notes: regstrength is the Democratic Community variable developed in 
. * Kadera, Crescenzi & Shannon, AJPS 2003. 
. 
. *cannot run this with G&W's "diff1" variable, due to multicollinearity
. corr regstrength diff1
(obs=186)

             | regstr~h    diff1
-------------+------------------
 regstrength |   1.0000
       diff1 |   0.7699   1.0000


. *correlated at 0.7699* 
. 
. nbreg fat1 regstrength pcenerg states year, nolog robust

Negative binomial regression                    Number of obs     =        186
                                                Wald chi2(4)      =     199.60
Dispersion           = mean                     Prob > chi2       =     0.0000
Log pseudolikelihood = -284.59663               Pseudo R2         =     0.2214

------------------------------------------------------------------------------
             |               Robust
        fat1 |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
 regstrength |  -.1274252    .053537    -2.38   0.017    -.2323558   -.0224947
     pcenerg |  -.8749437   .3854213    -2.27   0.023    -1.630355   -.1195318
      states |  -.0013405   .0034654    -0.39   0.699    -.0081324    .0054515
        year |   .0367342   .0060307     6.09   0.000     .0249143     .048554
       _cons |  -68.90119   11.25085    -6.12   0.000    -90.95244   -46.84993
-------------+----------------------------------------------------------------
    /lnalpha |  -2.591516    .745436                     -4.052544   -1.130489
-------------+----------------------------------------------------------------
       alpha |   .0749064   .0558379                      .0173781    .3228755
------------------------------------------------------------------------------

. 
. *MODEL E, Adding nonlinear term
. nbreg fat1 c_regstrength c_regstrsq pcenerg states year, nolog robust

Negative binomial regression                    Number of obs     =        186
                                                Wald chi2(5)      =     198.48
Dispersion           = mean                     Prob > chi2       =     0.0000
Log pseudolikelihood = -284.50172               Pseudo R2         =     0.2217

-------------------------------------------------------------------------------
              |               Robust
         fat1 |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
c_regstrength |  -.1144374   .0553027    -2.07   0.039    -.2228287   -.0060461
   c_regstrsq |   -.004921   .0145172    -0.34   0.735    -.0333743    .0235323
      pcenerg |  -.8704781   .3815371    -2.28   0.023    -1.618277   -.1226791
       states |  -.0005438   .0042539    -0.13   0.898    -.0088814    .0077937
         year |    .035342   .0062723     5.63   0.000     .0230485    .0476356
        _cons |   -66.2956   11.74114    -5.65   0.000    -89.30781    -43.2834
--------------+----------------------------------------------------------------
     /lnalpha |  -2.629525   .7919085                     -4.181637   -1.077413
--------------+----------------------------------------------------------------
        alpha |   .0721127   .0571067                      .0152735    .3404752
-------------------------------------------------------------------------------

. 
. *Here is a version with systemic trade, which reduces the N to 113.*
. *change in sample size is due to trade data starting later.*
. 
. *MODEL F: adding trade
. nbreg fat1 regstrength pcenerg sysdepstate states year, nolog robust

Negative binomial regression                    Number of obs     =        113
                                                Wald chi2(5)      =     115.16
Dispersion           = mean                     Prob > chi2       =     0.0000
Log pseudolikelihood = -211.79609               Pseudo R2         =     0.1602

------------------------------------------------------------------------------
             |               Robust
        fat1 |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
 regstrength |  -.1359357   .0604162    -2.25   0.024    -.2543493    -.017522
     pcenerg |  -.8583054    .426128    -2.01   0.044    -1.693501   -.0231099
 sysdepstate |  -27.75942   19.30681    -1.44   0.150    -65.60007    10.08124
      states |  -.0015185   .0049733    -0.31   0.760     -.011266     .008229
        year |   .0371273   .0072786     5.10   0.000     .0228615    .0513931
       _cons |  -69.55923    13.7169    -5.07   0.000    -96.44387    -42.6746
-------------+----------------------------------------------------------------
    /lnalpha |  -2.415045   .6648423                     -3.718112   -1.111978
-------------+----------------------------------------------------------------
       alpha |   .0893633   .0594125                      .0242798    .3289077
------------------------------------------------------------------------------

. 
. 
. ** Replication of Table 3 is done with a different dataset
. 
. 
. 
. 
. 
. 
. 
end of do-file

. log close
      name:  <unnamed>
       log:  /Users/Mark 1/Dropbox/Grant Proposals ICES/ISQ Response/Accepted/replication/replicating table 2.log
  log type:  text
 closed on:  10 Jun 2015, 16:53:00
-----------------------------------------------------------------------------------------------------------------
